Update Dockerfile

Dockerfile

@@ -1,37 +1,23 @@
-# Use an official NVIDIA CUDA image that includes the full CUDA Toolkit (nvcc).
-# This is required to compile flash_attn from source.
-# We are using CUDA 12.1.1 which is compatible with torch 2.1.2+cu121.
-FROM nvidia/cuda:12.1.1-devel-ubuntu22.04
-
-# Avoid prompts during package installation
-ENV DEBIAN_FRONTEND=noninteractive
-
-# Install Python 3.9 and pip
-RUN apt-get update && \
-    apt-get install -y python3.9 python3.9-venv python3-pip python3.9-dev && \
-    apt-get clean
-
-# Create a symlink so 'python' and 'pip' point to the correct versions
-RUN ln -s /usr/bin/python3.9 /usr/local/bin/python
-RUN ln -s /usr/bin/pip3 /usr/local/bin/pip
+# Use an official Python runtime as a parent image.
+FROM python:3.9-slim
 
 # Create a dedicated, non-root user for security.
 RUN useradd -m -u 1000 user
 USER user
 
-# Set environment variables for the new user
+# Set environment variables
 ENV HOME=/home/user
 ENV PATH=$HOME/.local/bin:$PATH
 ENV TRANSFORMERS_CACHE=$HOME/.cache
 WORKDIR $HOME/app
 
-# Copy requirements file and install dependencies
+# Copy and install requirements
 COPY --chown=user:user ./requirements.txt .
 RUN pip install --no-cache-dir -r requirements.txt
 
-# Copy the rest of the application
+# Copy the application file
 COPY --chown=user:user ./app.py .
 
-# Expose the port and run the application
+# Expose the port and run the app
 EXPOSE 7860
 CMD ["uvicorn", "app:app", "--host", "0.0.0.0", "--port", "7860"]